When the Trump administration decided its southern border crisis was partly a foreign policy problem, it went to the State Department, not just Homeland Security.

It suspended aid to El Salvador, Guatemala, and Honduras until those governments took measurable steps to stop northward migration. It pushed asylum cooperation agreements that made Central American states process claims before migrants ever reached United States (US) soil. It threatened Mexico with tariffs until Mexican troops moved to the border. When Colombia turned back two US deportation flights in January 2025, Washington ordered 25% tariffs, a travel ban, and visa sanctions the same day, and Bogotá accepted every US term within hours. It has since struck deportation deals with more than two dozen governments.

Whatever one thinks of the ethics of that approach, it rests on the premise that a country with leverage over the states its undocumented migrants come from can use that leverage, and refusing to use it is itself a policy choice.

There are several techniques used in the taxonomy of statecraft in altering an environment, ranging from propaganda to diplomacy to military. Let us concern ourselves with economic statecraft, which refers to influence attempts relying on resources that carry a market price, things such as trade access, aid flows, investment decisions, and institutional financing. Influence of this kind rests on the ability to offer rewards for compliance and impose costs for non-compliance.

States that attempt to exert influence in the world solely on the basis of reasoned argument are likely to be disappointed. The capacity to make effective threats or promises is generally known in advance by interested parties and does not have to be made explicit. Washington’s tariff threat to Colombia was effective precisely because it made explicit what was already known implicitly; that the US possessed asymmetric leverage and was willing to use it.

South Africa possesses its own version of this leverage but has structured its foreign policy as though it does not.

Foreigners in SA

The 2022 Census from Statistics South Africa (Stats SA) counted 2.4 million international migrants in South Africa, 86% of them from the Southern Africa region and just under half from Zimbabwe alone. Stats SA’s 2018 working estimate was closer to four million international migrants, and the Census carried a 31% undercount.

Although the exact number moves, the pattern since 1996 has not. The people arriving come overwhelmingly from countries the African National Congress (ANC) has spent thirty years declining to publicly challenge.

South Africa entered the post-1994 order promising liberal internationalism, but instead let a liberationist solidarity with fraternal ruling parties crowd it out on every issue where the two conflict. South Africa tries to hold three roles at once: regional leader, mediator, liberation supporter. The first two demand the normative independence to call out conduct that violates the rules South Africa claims to uphold. The third demands the opposite, a fraternal posture toward parties that fought alongside the ANC. When these roles collide, the liberation solidarity posture always wins.

South Africa’s main tool in Southern Africa has been diplomacy-based soft power, but this soft power operates without the material incentives that give it teeth. South Africa has no carrots to offer. There is no reward for improved governance among regional states, no cost for the absence of it.

The concept of soft power, properly understood, requires a hegemon that can incentivise another state to act in a particular way. Pretoria has relied instead on persuasive diplomacy, a weaker instrument that depends on goodwill rather than consequence.

This is consistent with what South African foreign policy scholars describe as hegemonic stability theory. The hegemon provides public goods and sets the rules of the regional system, not as an act of altruism but on the basis of enlightened, medium-term self-interest. The country with the largest stake in the preservation of the system bears the transaction costs of running it because the alternative, disorder, is more expensive. But the theory also warns that hegemonic leadership is a wasting asset.

When the hegemon begins to be perceived by other members of the system as acting solely in its own interests, or, as in South Africa’s case, as failing to act at all, it loses the legitimacy to shape outcomes. South Africa has accumulated the institutional markers of regional leadership: the Southern African Development Community (SADC), the African Union chair rotation, the G20, BRICS, near-permanent circulation through the United Nations Security Council. But institutional position without the exercise of substantive influence is merely symbolic hegemony.

Failure to Evolve

This posture has evolved, or failed to evolve, across presidential administrations. Under Nelson Mandela, foreign policy was idealist in aspiration but lacked direction; the Nigerian crisis under Sani Abacha taught the ANC the political cost of publicly confronting a fellow African government.

Under Thabo Mbeki, the African Renaissance philosophy produced the most active period of South African regional engagement, but its signature diplomatic instrument, quiet diplomacy, was deployed most conspicuously toward Zimbabwe.

Under Jacob Zuma, regional engagement contracted sharply in favour of BRICS and bilateral ties with China.

Under Cyril Ramaphosa, the orientation has been economic diplomacy, focused on trade and value chains. None of these administrations has treated the link between regional governance and South African immigration as a policy problem that foreign affairs should own.

Mbeki’s quiet diplomacy through Zimbabwe’s 2008 election, which Morgan Tsvangirai won and Robert Mugabe kept anyway, set the template. SADC observer missions have rated Zimbabwe’s 2018 and 2023 elections more gently than other international observers. Pretoria’s response to Mozambique’s violent 2024 post-election crackdown was muted. South African troops sat in the eastern Democratic Republic of the Congo under SADC’s peacekeeping mission in that country, while the political conditionality that might shape Kinshasa’s conduct never reached the table. Lesotho’s recurring political crises get the SADC ministerial circuit.

Mugabe destroyed Zimbabwe, not Pretoria. The governance failures of the Front for the Liberation of Mozambique (Frelimo), the governing party there, are Frelimo’s. South African criticism alone would not have changed any of these outcomes. All this is true, but it is beside the point.

Washington’s aid suspensions indeed did not single-handedly redesign Central American governance. Leverage does not have to guarantee an outcome to be worth using, and declining to use it is not an expression of neutrality or non-alignment.

Sensitive

South Africa is highly sensitive to governance failures in the region. Sensitivity measures how quickly a state feels the effects of changes in a given issue-area, while vulnerability measures the extent to which that state can control its response. When these states deteriorate, people move to South Africa, and the effects register immediately in the labour market, in social services, in political tension. But South Africa is not helpless and has options it has chosen not to exercise.

Sensitivity without vulnerability is uncomfortable but manageable, while sensitivity with high vulnerability presents a high-risk exposure. Pretoria has mistaken a choice to absorb the costs for inevitability.

Moreover, states regularly employ strengths in one issue-area to compensate for weaknesses in another. The 1971 Smithsonian Crisis, where the US used trade sanctions and security guarantees as bargaining chips to force currency realignments, linked issue-areas that were formally supposed to be kept separate.

Trump’s approach to Central American migration is the same logic applied more crudely; linking aid, trade access, and visa privileges to migration enforcement. Pretoria has all the ingredients for its own version of this cross-issue linkage, but has never attempted it.

South Africa’s economy accounts for roughly 45% of the aggregate SADC GDP. Southern Africa receives 86% of South African intra-Africa exports, and South African businesses, from Shoprite and MTN to Standard Bank and Eskom, have made deep inroads across the region. South African construction firms, mining companies, telecommunications operators, and financial institutions are present in nearly every SADC member state.

This commercial footprint creates a dependency relationship that has never been harnessed for governance objectives. There are numerous South African businesses located in Southern African countries, yet there are virtually no businesses from those countries with significant footprints in South Africa. The asymmetry is tangible, and it gives Pretoria leverage it has never acknowledged, let alone deployed.

Aid

South Africa also runs a direct aid budget pointed at the region. Between 2017 and 2024 the African Renaissance Fund (ARF) paid out to the very neighbours the migrants come from: R42 million in cyclone relief to Zimbabwe, R34 million for agricultural recovery in Mozambique’s Cabo Delgado, R40 million in food aid to Eswatini, and R14 million to steer Lesotho’s reform process. It funds the SADC election observer missions too, the ones that then rate Zimbabwe’s votes more gently than anyone else does. The money goes in as disaster relief and quiet facilitation, but is never tied to the governance conduct.

Neoliberal institutionalists argue that states cooperate when cooperation produces absolute gains. As long as all parties benefit, unequal distribution is tolerable. Neorealists counter that states focus on relative gains, scanning each other for signs that the distribution of benefits is shifting against them. South Africa’s regional posture has been almost entirely on the neoliberal side of this divide.

Pretoria has provided public goods (infrastructure, aid, peacekeeping) and absorbed costs (migration, trade deficits with weaker states) without asking whether the relative distribution of gains serves its interests.

The mechanism that prevents free-riding, the hegemon’s willingness to enforce the rules, is exactly what South Africa refuses to activate. There is a reason for this refusal that goes deeper than policy inertia. Leverage is not a strategic option the Department of International Relations and Cooperation (DIRCO) can simply pick up. The ANC’s posture toward the governing parties in Mozambique, Angola, Zimbabwe, and Namibia runs through the party’s understanding of itself, not through a policy document a new foreign minister could revise.

South Africa’s foreign policy has been president-driven, shaped by the ANC’s own identity and its historical bonds with liberation movements across the region. The party’s reluctance to confront fraternal ruling parties is not a diplomatic calculation; it is an identity commitment. That is why the Government of National Unity (GNU) has produced no discernible shift here despite the Democratic Alliance holding cabinet seats that touch foreign affairs. South African foreign policy on this question runs through the ANC’s identity, not through DIRCO.

Not Fixed

Constructivists argue that state identities are not fixed; they are produced and reproduced through practice. This implication is that the ANC’s liberation solidarity posture is not a permanent structural feature but a constructed identity that could, in principle, be reconstructed if the political conditions changed. But the same framework also warns that agent-structure relationships are sticky. Just because a structure is the product of human agency does not mean it will be easy for human agents to change its nature once it has been established.

The ANC’s liberation-era bonds are deeply institutionalised in party culture, diplomatic practice, and multilateral positioning. Reconstructing them would require a political shock of a kind the GNU has not delivered.

What Washington’s approach demonstrates, however crudely, is that a state’s leverage over its migrant-sending neighbours is a policy instrument to be considered. However, the conditional friendship Washington applied coercively is not the only model. Pretoria does not need Trump-style diplomacy to test the underlying premise. It could start with a hegemonic posture that uses institutional frameworks and economic incentives to shape regional behaviour, rather than threats and penalties.

It could tie ARF disbursements to measurable governance benchmarks. It could condition project financing on compliance with SADC’s own stated principles of democratic governance and human rights.

None of this requires coercion, it requires honesty about a relationship that currently operates on unconditional terms. The cheapest border policy is a foreign policy that expects something back from its friends. Until South Africa adopts this, the immigration debate will stay at Home Affairs, arguing about a symptom whose cause sits three ministries away.

Ofentse Davhie is a research associate at the Centre for Risk Analysis with a focus on political risk and foreign policy.